GenDa - Simulated Processing Times

A synthetic dataset consisting of 5,000 mortgage application records . These records were designed to mirror real-world heterogeneity in applicant demographics and process complexity. Each application was assigned attributes such as applicant age, income, credit score, employment status, loan purpose, documentation completeness, data quality, and complexity. Stepwise process times were constructed from the industry-validated estimates, to simulate both the legacy(baseline) and proposed GenDa workflows本合成数据集包含5000条抵押贷款申请记录。此类记录旨在还原申请人人口统计特征与审批流程复杂度的真实异质性。每份申请均包含以下属性：申请人年龄、收入水平、信用评分、就业状态、贷款用途、申请材料完整性、数据质量以及流程复杂度。逐步流程耗时基于行业验证的估算值构建，用于模拟传统（基线）与拟议的GenDa工作流。